sequent, not the cause, of physical expression, though the whole discussion was permeated with the implication that it was either an antecedent or a concomitant of such an expression. This appeared paradoxical and absurd to most students, because it described what is true of reflexive emotion, and what is not true of the emotions Professor James was discussing. Pleasure and pain are consequent accompaniments of exercised functions, but the "motor" consciousness is not a consequent of the physical expression it engenders, and the "emotion" that follows this expression is pleasure or pain. What is required, therefore, is an analysis and classification of mental phenomena which will give a definite conception to the terms 'feeling' 'emotion,' 'sensibility,' 'passion,' etc., or afford a substitute for them.
Now if we are to accept the tripartite division of mental phenomena, we must use the term 'emotion' in one of two senses, either a broader or a narrower. The most comprehensive meaning must be one which will include the sensibilities and the passions. The common characteristics of excitement, tone, or intensity, which may be supposed to describe both classes, may not be sufficient to justify their classification under the same head, and, if so, we are left to the narrower denotation of the term in which it must apply either to the reflexes of function or to the "motor" tendencies of consciousness. But however we use the term, we must radically distinguish between the reflexive aspect of 'emotion,' as I choose to call it, which is an incident of functional and other action, and what I shall call the impulsive aspect, which is a dynamic function of consciousness. The former is a consequent, and the latter an antecedent of action. The difference between them is the difference between the pleasure of music as an 'emotion' and anger as an 'emotion.' Impulsive 'emotion,' at least, lies on the borderland of the will, if it is not inceptive will itself. But if it be regarded as in its nature belonging to the will, though not in its deliberative form, we can sustain the tripartite division only by limiting the term 'emotion' or 'feeling' to pleasure and pain. Otherwise we should have either to choose the abstract and general import, in which the term would cover two unique classes of mental states, or to decide upon a fourfold division: intellection, emotion (pleasure, pain), passion, and conation. Perhaps the latter division could be reduced to a twofold one,—the old classification, intellectual and active powers, with two separate classes of accidents which are each mere modifications of their respective principals. This is in effect, of course, a quadripartite classification.
